marketing-china-market-localization-strategist

Pass

Audited by Gen Agent Trust Hub on Aug 10, 2026

Risk Level: SAFEPROMPT_INJECTIONNO_CODE
Full Analysis
  • [PROMPT_INJECTION]: The skill instructs the agent to analyze untrusted third-party data from social media platforms, creating a potential surface for indirect prompt injection.
  • Ingestion points: The workflow involves aggregating hotlist data and performing a 'Comment Track' analysis across platforms like Douyin, Xiaohongshu, and Weibo (SKILL.md).
  • Boundary markers: The instructions lack specific markers or requirements to delimit external input or to ignore instructions embedded within social media content.
  • Capability inventory: The skill requires capabilities for monitoring external platforms and generating content or strategies based on that data.
  • Sanitization: No sanitization or filtering procedures are defined for handling instructions hidden within external trend signals.
  • [NO_CODE]: The skill consists entirely of instructional markdown and does not include any scripts or configuration files that execute code.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 10, 2026, 02:37 AM
Security Audit — agent-trust-hub — marketing-china-market-localization-strategist